Datadog’s big bark: why AI may not be the end of software

Key takeaways

• Datadog’s strong results show AI can boost some software demand, not only disrupt it.

• The market is separating useful, embedded software from tools that AI may replace or bundle.

• For investors, the key question is simple: does AI make the product more necessary?

Datadog just gave the software sector something it badly needed: evidence that artificial intelligence (AI) is not only a threat. Sometimes, it is also a customer.

On 7 May 2026, Datadog closed at 188.73 USD up 31.3%, after the company reported stronger than expected first-quarter results and raised its full-year outlook. That is not a normal Thursday for a software stock. That is a labrador hearing the word “walk”.

Datadog sells cloud monitoring and security software. In plain English, it helps companies see what is happening inside their digital systems, spot problems, and fix them before customers notice. This is called observability, which sounds like a university word, but really means “do we know what just broke, where, and why?”

The bigger message goes beyond one company. For months, investors have worried that AI will hurt software-as-a-service (SaaS), where customers rent software online instead of buying it once. The fear is simple: if AI agents can write code, handle support, analyse data and automate office tasks, some software tools may become less valuable. Datadog’s results show the story is more nuanced. AI is not taking a flamethrower to software. It is sorting the useful from the replaceable.

The market liked growth with a reason

Datadog’s first-quarter revenue rose 32% from a year earlier to 1.01 billion USD. The company also lifted its 2026 revenue outlook to between 4.30 billion USD and 4.34 billion USD, above earlier guidance and analyst expectations. Adjusted earnings were 0.60 USD per share, also ahead of expectations, according to estimates compiled by Bloomberg.

The quality of the growth mattered. Datadog ended the quarter with about 4,550 customers spending at least 100,000 USD in annual recurring revenue, up 21% from a year earlier. Annual recurring revenue means the revenue a company expects to repeat over a year from subscriptions. For investors, this helps show whether customers are staying and expanding, not just signing one-off deals.

The company also generated 289 million USD of free cash flow. Free cash flow is the cash left after running and investing in the business. It matters because growth funded by real cash is usually healthier than growth funded mainly by hope, conference slides and very expensive coffee.

This is why the share price reaction was so strong. Investors were not only reacting to a beat. They were reacting to a company that appears to sit where two forces meet: cloud complexity and AI complexity.

AI is not killing software. It is changing the test

The old SaaS question was: how many people use the software? The new AI question is: how deeply does the software sit inside the customer’s operations?

That is a big difference. Tools that charge per employee may face pressure if AI reduces headcount or automates tasks. Simple workflow tools may also face bundling risk, where large platforms such as Microsoft, Google or Salesforce add similar features inside broader packages. When a feature becomes part of the furniture, standalone pricing can get harder.

Datadog is a different type of software company. It is closer to digital infrastructure. When companies use more cloud services, more applications, more AI models and more automated agents, their systems become harder to manage. More moving parts mean more things can fail. Datadog sells the control room.

AI may therefore increase the need for observability. Large language models, the systems behind tools such as chatbots and agents, can fail in ways traditional software did not. They can slow down, produce errors, overload systems, or cost too much to run. Companies using AI need to know what is happening across models, graphics processing units (GPUs), data flows, security alerts and customer experiences.

That is the important lesson. AI does not treat all software equally. It pressures some tools, lifts others, and forces every company to prove its relevance.

From selling seats to solving problems

The SaaS model is also changing. Many older software businesses grew by selling more seats, meaning more users inside a customer. That model works well when employment grows and software spreads across departments. It looks less bulletproof when AI agents can do more work with fewer human clicks.

Datadog’s model benefits more from usage and complexity. If a customer runs more applications, produces more data, or deploys more AI systems, the need for monitoring can rise. This does not make the business risk-free, but it gives investors a clearer reason why AI can be a tailwind.

That distinction is useful across the software sector. A company may be more resilient if it helps customers manage risk, lower costs, protect data, improve uptime, or run AI safely. A company may be more exposed if AI can easily copy its output, reduce its user base, or allow a larger platform to bundle the same function.

In short, software investors now need a sharper filter. “AI exposure” is not enough. Every company claims that, preferably before lunch. The better question is whether AI makes the product more necessary, less necessary, or merely more marketable.

Risks: not every bark means a moat

There are still risks. Datadog’s valuation now reflects a lot of good news, so even small disappointments could matter. Strong growth can support high expectations, but high expectations rarely come with a safety helmet.

Competition is another risk. Cloud providers, security vendors and other software platforms all want more of the same budget. Customers may also try to simplify their software stacks, especially if the economy weakens.

Finally, AI demand itself could become uneven. Companies are still learning how much value they get from AI tools in production. Watch for signs such as slower customer growth, weaker usage expansion, lower free cash flow margins, or management talking more about “long-term opportunity” than near-term adoption.

Investor playbook: follow the usefulness

• Separate AI labels from AI economics. Look for revenue growth, cash flow and customer expansion.
• Watch pricing models. Usage-based revenue may benefit when digital activity grows.
• Check whether AI raises complexity. More complexity can support monitoring, security and data tools.
• Keep position size in context. Great stories can still become expensive stories.

The bigger lesson

Datadog’s results do not prove that all software is safe from AI disruption. They prove something more useful: the software sector is not one single story. AI can replace some tasks, weaken some pricing models, and make some tools look ordinary. But it can also create new problems that companies must monitor, secure and control.

Datadog rallied because investors saw a software business standing near that control layer. For long-term investors, that is the real takeaway. The AI age may not reward every software company, but it still needs software that keeps the machines from turning the office into a very expensive guessing game.

This material is marketing content and should not be regarded as investment advice. Trading financial instruments carries risks and historic performance is not a guarantee of future results.